**1. Introduction**

Wall-mounted room air conditioners were invented in Japan and spread mainly as standard household air conditioners in East Asia. These are split-type air conditioners where the indoor and outdoor units are separate. The indoor unit uses a cross-flow fan, which reduces the depth and length of the unit and allows it to be mounted on a wall. The ability to install the air conditioner on the wall has the advantage of reducing the floor space required for installation. Design problems in indoor units are (1) the

fin-and-tube heat exchanger design and (2) the airflow control. In the following sections, we introduce two numerical models for development of indoor units.

#### **1.1 Fin-and-tube heat exchanger model**

**Figure 1a** is a photo of an indoor unit **Figure 1b** illustrates a sectional view of the internal structure of the indoor unit. The indoor unit consists of three parts: the finand-tubes heat exchanger, the cross-flow fan, and the vanes. The internal airflow field pattern can be explained with a two-dimensional sectional view [1, 2]. The fan supplies air for the heat exchanger from the air inlet, the heat exchanger raises the air temperature, and then the vanes control the airflow direction, which enhance the air conditioner's thermal comfort and energy-saving performance.

The fin-and-tube heat exchanger consists of refrigerant tubes inserted into fins. Each tube can be connected with a U-shaped bent tube to form a refrigerant flow circuit with an inlet and outlet for the refrigerant flow (**Figure 1c**). A refrigerant flow circuit has a refrigerant inlet and outlet, and the single continuous tube connecting each is called the "path."

An important improvement for heat exchanger performance is to investigate the air and refrigerant flow in the heat exchanger, identify areas that are causing energy losses that do not contribute to heat exchange between the air and refrigerant flow, and improve those areas.

A main objective of the fin-and-tube heat exchanger model development is to quantify the refrigerant flow state such as pressure, enthalpy, and temperature in wall-mounted indoor unit heat exchangers, especially those units equipped with cross-flow fans.

In previous research papers [3, 4], the heat exchanger was modeled as a straightforward shape (**Figure 2a**). The velocity and direction of the airflow in the heat exchanger were assumed to be unchanged, so that the airflow volume distribution was given in the first row of the tube array and the second and third rows were simply the same as the distribution in the first row. These assumptions allow us to solve for enthalpy of refrigerant flow, pressure, mass flow rate, and heat transfer between the refrigerant flow and air, even when the heat exchanger has multiple paths connected by a distributor device. The previous study [5, 6] developed indoor unit CFD models, which can predict fin pressure loss and heat transfer of fins. These models do not include refrigerant flow calculation, therefore, the transition between the pressure and entropy values from the refrigerant pipe inlet to the outlet in the path design was unknown. The previous study [7] developed an inclined heat exchanger CFD model

#### **Figure 1.**

*Indoor unit: (a) a photo of an indoor unit, (b) internal structure of indoor unit, and (c) a structure of heat exchanger.*

*Application of CFD to Prediction of Heat Exchanger Temperature and Indoor Airflow Control… DOI: http://dx.doi.org/10.5772/intechopen.110076*

#### **Figure 2.**

*Heat exchanger modeling: (a) a schematic of previous research model [3, 4], (b) an indoor unit of fin-and-tube heat exchanger, and (c) streamline of indoor unit heat exchanger.*

#### **Figure 3.**

*Indoor airflow control: (a) short airflow circulation and (b) long airflow circulation.*

and calculate airflow distribution. It incorporates air flow distribution to refrigerant circuit calculation model, however, this model can reflect indoor unit heat exchanger.

The heat exchanger actually used in the indoor unit is not a straight type but a folded type (**Figure 2b**) and airflow streamline is not a straight but curved form (**Figure 2c**). Since airflow does not flow straight through the first row of tube arrays, it is difficult to determine the air flow rate distribution input to the heat exchanger. Furthermore, the airflow volume distribution in the first row of the tube array is not the same as the tube array in the other rows of the actual heat exchanger, as assumed in previous research papers. Our numerical model is coupled with CFD for air-side calculations and with the one-dimensional refrigerant flow circuit model for refrigerant-side calculations. This CFD model reflects an airflow distribution in the fins and can calculate the pressure and entropy values of the refrigerant flow.

#### **1.2 Indoor airflow model**

**Figure 3** shows an example of indoor airflow control. A wall-mounted air conditioner was installed on the wall and positioned to blow warm air from a high position in the room. The distance of the blowing air can be varied by controlling the fan speed, heat exchanger temperature, and vane angle. **Figure 3a** is a schematic diagram of short circulation control. The red line represents the airflow moving away from the indoor unit, and the blue line represents the airflow approaching the indoor unit. For explanation, we refer to the red line airflow as the blowing airflow and the blue line airflow as suction airflow. When the vanes are angled upward, the airflow is directed upward. In the heating mode, the temperature of blowing airflow is higher than the air temperature in the room. The blowing airflow floats upward because a buoyant force

works. The airflow circulates near the indoor unit, and the reach of the airflow of the tip is short.

**Figure 3b** shows a schematic diagram of long circulation control. When the vane is angled down, the airflow is directed downward. Because the suction force is weak, blowing airflow tends to travel farther from the indoor unit and has a long circulatory path.

The airflow control is a technology that sends hot or cold air to various positions. To improve energy-saving performance and comfort, it is optimal to control airflow factors, for example, by avoiding airflow to areas of the room where the user is not present and selectively applying more airflow to body parts that contribute significantly to comfort, such as the feet [8].

If such airflow control can be simulated numerically, the work required for testing the actual equipment can be reduced. The objective of developing indoor airflow model is to replace testing with numerical calculations and to quantitatively evaluate wind velocity and temperature distributions, which are difficult to measure in testing.

Although there have been many studies using CFD in indoor airflow [9], few studies have dealt with split-type air conditioners. Casado et al. [10] verified the airflow temperature distribution using a split-type air conditioner. In this study, the vane angle was fixed. The case where the vane angle and wind direction are varied is verified by Lee S et al. [11]. The numerical model was validated using experimental data under isothermal conditions by varying the vane angle and wind direction. The numerical model was validated using experimental data under non-isothermal conditions with varying vane angles and airflow directions. Experimental data under nonisothermal conditions were used to validate the numerical model. Temperature distributions were obtained when the air conditioner was operated in heating mode and the vane angle and airflow rate were varied.
